天气预报论文趋势 2023–2026

主要结论

• 天气预报论文从 2023 年的 26 篇增长到 2026 年的 158 篇，三年间扩大到约 6.1 倍。
• NeurIPS 是发天气预报论文最多的会议，2023–2026 年共 137 篇，占四个会议总量的 38%。
• 交叉方向里涨得最快的是扩散模型：同时属于两个主题的论文从 2023 年的 1 篇增至 2026 年的 13 篇。
• 在「气候与物理科学」的 5 个主题里，它的 2026 年论文数排第 2；同组增长最快的科学仿真是 4.3 倍，天气预报是 6.1 倍。

代表论文

按年份和会议均匀挑选，方便快速看到每一届的样貌

• NeurIPS 2026AdaST: Adaptive Coupling for Spatial-Temporal Forecasting
• ICML 2026Doubly Outlier-Robust Online Infinite Hidden Markov Model
• ICLR 2026ARFlow: Auto-regressive Optical Flow Estimation for Arbitrary-Length Videos via Progressive Next-Frame Forecasting
• CVPR 2026NimbusGS Unified 3D Scene Reconstruction under Hybrid Weather
• NeurIPS 2025DAWP: A framework for global observation forecasting via Data Assimilation and Weather Prediction in satellite observation space
• ICML 2025Information Bottleneck-guided MLPs for Robust Spatial-temporal Forecasting
• ICLR 2025CirT: Global Subseasonal-to-Seasonal Forecasting with Geometry-inspired Transformer
• CVPR 2025Continuous Adverse Weather Removal via Degradation-Aware Distillation
• NeurIPS 2024Off to new Shores: A Dataset & Benchmark for (near-)coastal Flood Inundation Forecasting
• ICML 2024Conformalized Adaptive Forecasting of Heterogeneous Trajectories
• ICLR 2024ClimODE: Climate and Weather Forecasting with Physics-informed Neural ODEs
• CVPR 2024HPNet: Dynamic Trajectory Forecasting with Historical Prediction Attention
• NeurIPS 2023SubseasonalClimateUSA: A Dataset for Subseasonal Forecasting and Benchmarking
• ICML 2023ClimaX: A foundation model for weather and climate
• ICLR 2023Compressing multidimensional weather and climate data into neural networks
• CVPR 2023Trajectory-Aware Body Interaction Transformer for Multi-Person Pose Forecasting
• NeurIPS 2026AirMPA: A Meteorology-to-Pollution Adapter for Global Air Quality Forecasting
• ICML 2026Lightweight and Interpretable Transformer via Unrolling of Mixed Graph Algorithms for Traffic Forecast
• ICLR 2026TianQuan-S2S: A Subseasonal-to-Seasonal Global Weather Model via Incorporate Climatology State
• CVPR 2026STCast Adaptive Boundary Alignment for Global and Regional Weather Forecasting

全部天气预报论文

共 158 篇，第 2 / 4 页

51. ICML 2026Beyond Model Ranking: Predictability-Aligned Evaluation for Time Series Forecasting
52. ICML 2026Byte Pair Encoding for Efficient Time Series Forecasting
53. ICML 2026Channel Adapter for Time Series Foundation Models in Zero-Shot Multivariate Forecasting
54. ICML 2026ChaosNexus: A Foundation Model for ODE-based Chaotic System Forecasting with Hierarchical Multi-scale Awareness
55. ICML 2026ClimateAR: Multi-Scale Autoregressive Generative Modeling for Climate Forecasting
56. ICML 2026CombinationTS: A Modular Framework for Understanding Time-Series Forecasting Models
57. ICML 2026Crisp: A Spectral-Based Interaction Strategy for Multivariate Time Series Forecasting
58. ICML 2026DAG: A Dual Correlation Network for Time Series Forecasting with Exogenous Variables
59. ICML 2026Doubly Outlier-Robust Online Infinite Hidden Markov Model
60. ICML 2026DropoutTS: Sample-Adaptive Dropout for Robust Time Series Forecasting
61. ICML 2026DSENet: A Novel Dual-Stream Enhancement Network for Multi-Scale Non-Stationary Time Series Forecasting
62. ICML 2026Ellipsoidal Time Series Forecasting
63. ICML 2026EMFormer: Efficient Multi-Scale Transformer for Accumulative Context Weather Forecasting
64. ICML 2026FIPN: Forward Self-Organizing Interpretable Polynomial Networks for Time Series Forecasting
65. ICML 2026FlowState: Sampling-Rate‑Equivariant Time‑Series Forecasting
66. ICML 2026From Observations to States: Latent Time Series Forecasting
67. ICML 2026From Text to Forecasts: Bridging Modality Gap with Temporal Evolution Semantic Space
68. ICML 2026FutureOmni: Evaluating Future Forecasting from Omni-Modal Context for Multimodal LLMs
69. ICML 2026Global Merger-Arbitrage Forecasting with Language Models
70. ICML 2026HVR-Met: A Hypothesis-Verification-Replanning Agentic System for Extreme Weather Diagnosis
71. ICML 2026Information Geometry Loss for Time Series Forecasting
72. ICML 2026Invariant Representation Learning for Source-Free Time Series Forecasting with LLM-Centric Proxy Denoising
73. ICML 2026L-Drive: Beyond a Single Mapping—Latent Context Drives Time Series Forecasting
74. ICML 2026LagLLM: LLM-empowered lead–lag dependency learning for spatial-temporal time series forecasting
75. ICML 2026Lightweight and Interpretable Transformer via Unrolling of Mixed Graph Algorithms for Traffic Forecast
76. ICML 2026MemCast: Memory-Driven Time Series Forecasting with Experience-Conditioned Reasoning
77. ICML 2026Nested Spatio-Temporal Time Series Forecasting
78. ICML 2026Once-for-All: Scalable Simultaneous Forecasting via Equilibrium State Estimation
79. ICML 2026Overcoming the Modality Gap in Context-Aided Forecasting
80. ICML 2026Parameter Decorrelation via Transition-Variance Alignment for Multivariate Time-series Forecasting
81. ICML 2026Parametric Prior Mapping Framework for Non-stationary Probabilistic Time Series Forecasting
82. ICML 2026PnP-Corrector: A Universal Correction Framework for Coupled Spatiotemporal Forecasting
83. ICML 2026PULSE: Generative Phase Evolution for Non-Stationary Time Series Forecasting
84. ICML 2026Reviving Error Correction in Modern Deep Time-Series Forecasting
85. ICML 2026See More, Forecast Better and Faster: Enhancing Time Series Foundation Models via Inference-Time Plug-and-Play Downsampling
86. ICML 2026SEER: Transformer-based Robust Time Series Forecasting via Automated Patch Enhancement and Replacement
87. ICML 2026StretchTime: Adaptive Time Series Forecasting via Symplectic Attention
88. ICML 2026Taming the Recent-Data Bias: Towards Robust Time Series Forecasting with Global Context
89. ICML 2026The Label Horizon Paradox: Rethinking Supervision Targets in Financial Forecasting
90. ICML 2026Time Series Forecasting Through the Lens of Dynamics
91. ICML 2026TimeSeed: Effective Time Series Forecasting with Sparse Endogenous Variables
92. ICML 2026Transforming Weather Data from Pixel to Latent Space
93. ICML 2026U-Cast: A Surprisingly Simple and Efficient Frontier Probabilistic AI Weather Forecaster
94. ICML 2026WeatherSyn: An Instruction Tuning MLLM For Weather Forecasting Report Generation
95. ICML 2026What if Tomorrow is the World Cup Final? Counterfactual Time Series Forecasting with Textual Conditions
96. ICML 2026What If We Let Forecasting Forget? A Sparse Bottleneck for Cross-Variable Dependencies
97. ICML 2026WIND: Weather Inverse Diffusion for Zero-Shot Atmospheric Modeling
98. NeurIPS 2026AdaST: Adaptive Coupling for Spatial-Temporal Forecasting
99. NeurIPS 2026AirMPA: A Meteorology-to-Pollution Adapter for Global Air Quality Forecasting
100. NeurIPS 2026AME-TS: Anchored Mixture-of-Experts for Time Series Forecasting

常见问题

2023–2026 年四大会议一共有多少篇天气预报论文？

共 356 篇，按年份分别是 2023 年 26 篇、2024 年 77 篇、2025 年 95 篇、2026 年 158 篇。

NeurIPS 2026 有多少篇天气预报论文？

NeurIPS 2026 有 61 篇天气预报论文，上一届是 37 篇。

ICLR 2026 有多少篇天气预报论文？

ICLR 2026 有 34 篇天气预报论文，上一届是 22 篇。

ICML 2026 有多少篇天气预报论文？

ICML 2026 有 52 篇天气预报论文，上一届是 29 篇。

CVPR 2026 有多少篇天气预报论文？

CVPR 2026 有 11 篇天气预报论文，上一届是 7 篇。

哪个会议的天气预报论文最多？

NeurIPS，2023–2026 年共 137 篇。